Inflammation has been implicated in the pathophysiology of several chronic diseases. These diseases increase in prevalence in older and in sedentary individuals, but evidence is emerging that exercise has anti-inflammatory effects (4,12). For example, those who are physically active or who become physically active have a reduction in biomarkers used to assess systemic inflammation (13,14). Higher levels of physical activity are associated with lower mitogen-stimulated inflammatory cytokine production and lowered serum C-reactive protein levels (4,13,14). Exercise training also lowered skeletal muscle inflammatory protein content and reduced adipokine production and macrophage infiltration into fat depots (2,6).
Despite the apparent whole body and tissue-specific anti-inflammatory effects of exercise, the underlying mechanism for these effects is not known. Thus, our recent efforts have been directed at elucidating a mechanism for our early observation that a high level of physical activity was associated with a reduced lipopolysaccharide (LPS)-stimulated inflammatory cytokine production 4 (Fig. 1). We have focused our efforts on Toll-like receptor (TLR) 4-the primary signaling receptor for LPS-and found that TLR4 is higher in physically inactive individuals but reduced by exercise training (4,12-14).
The study of TLRs is a promising new direction in the search for potential mechanisms to explain exercise's ability to dampen inflammation. The far-reaching effects of inflammation and inflammatory products make it tempting to speculate that the positive effects of physical activity on the disease state may be elicited, at least in part, by the anti-inflammatory properties endowed by regular exercise.
INFLAMMATION, CHRONIC DISEASE, AND ANTI-INFLAMMATORY INFLUENCE OF EXERCISE
Low-level systemic inflammation is emerging as an important risk factor in chronic disease. For example, a biomarker of systemic inflammation-C-reactive protein-is a promising new risk factor for cardiovascular disease risk. Obesity, which seems to exert some of its comorbid influence through inflammatory processes, provides another interesting example of the potential inflammatory influences on chronic diseases-with adipocytes (or the macrophages that infiltrate them) able to produce an array of inflammatory products that could contribute to chronic diseases such as type II diabetes (2,17).
Markers of inflammation generally increase with age (6), but in a recent study (13), we found that physical activity status had a greater impact on LPS-stimulated cytokine production than age (Fig. 2). A unique aspect of this study was that we controlled for and excluded a variety of disease states to minimize their effects on our findings. Specifically, we found that whole blood cultures from both old inactive and young inactive subjects were "hyperresponsive" to LPS compared with young active and old active subjects (13).
There is no disputing the fact that the incidence of chronic disease increases as we age, and the list of diseases with an inflammatory etiology is growing. The debilitating effects of poorly regulated inflammation are becoming clearer, and the ability to modulate inflammatory processes with physical activity seems to be of considerable importance.
Evidence is mounting that exercise training has anti-inflammatory effects. These effects have been shown to occur in several tissues including monocytes (4,14), skeletal muscle (6), and adipose tissue (2). We found that resistance exercise training or high levels of physical activity were associated with reduced in vitro inflammatory cytokine production and other markers of systemic inflammation (4,12-14). Resistance exercise training also lowered skeletal muscle TNF-α protein content and mRNA in frail elderly (age, 81 ± 1 yrs) (6).
A novel idea-that inflammation may also play a key role in training-induced adaptations-emerged from Gielen's laboratory in a study published in December of 2005 (5). Gielen's group found that training-induced decreases in skeletal muscle inducible nitric oxide synthase expression and inducible nitric oxide synthase protein content were negatively correlated with muscle cytochrome c oxidase activity in patients with coronary heart failure (ejection fraction, <25%). Their general conclusion was that exercise-induced reductions in inflammation permitted adaptations in metabolic function. Simply put, their supposition was that reduced inflammation must precede oxidative adaptation. Because the change in cytochrome oxidase was strongly correlated to changes in V˙O2max, blunting of inflammation putatively allowed aerobic adaptation in these patients with diseases. This is a bold and, as yet, an unreplicated concept. The authors acknowledged that skeletal muscle adaptations alone do not account for all the changes in aerobic capacity. Nevertheless, this study underscores the emerging roles of inflammation in a wide range of physiological mechanisms.
Our early research showed a blunted response of LPS-stimulated whole blood cultures obtained from trained older women after 10 wk of resistive exercise training (4). We were surprised that exercise training engendered these responses and sought a mechanism for these apparent exercise training-induced changes in monocyte function. Therefore, in our subsequent experiments, we measured TLR4 mRNA from the same group of trained women and compared the responses to age-matched sedentary controls. Our findings that TLR4 mRNA was lower in trained women compared with untrained women (4) led to our subsequent cross-sectional studies comparing cell-surface TLR4 expression in trained and untrained men and women (12,13) and a longitudinal training study of previously untrained men and women (14).
Toll receptors were first discovered in the fruit fly, Drosophila melanogaster, with 11 human homologues of toll-called TLRs-now identified (16). The TLRs are pattern recognition receptors with the ability to recognize pathogen-associated molecular patterns. Putative new endogenous ligands for TLRs are also emerging in the literature (16). A wide range of TLRs are available to respond to an array of pathogen-associated molecular patterns via intracellular signaling pathways and subsequent activation of the innate immune system by degrading the inhibitory protein inhibitor IκB, leading to mobilization of the transcription factor nuclear factor κβ (3). TLR4, the first TLR to be characterized and the focus of this review, is the primary signaling receptor for LPS-a component of gram-negative bacterial cell walls (3). TLR4 does not work alone but rather requires the assistance of the accessory signaling proteins, LPS-binding protein (LBP) and CD14-the so-called LPS receptor. As will be discussed, we speculate that endogenous products such as heat shock proteins (HSPs), that may be increased during or after exercise, could interact with and down-regulate TLR4 (9). It is also possible that exercise could influence the expression of LBP or CD14.
We found that subjects who are physically active or those who undertake resistive training lower mRNA and cell-surface expression of TLR4 (4,12-14). Despite its potential importance in the regulation of inflammatory processes, there are few exercise-related articles from which solid conclusions can be drawn (4,11,12,14). Nevertheless, exercise-induced blunting TLR expression is a promising area, which may explain the underlying effects of exercise on innate immune function and inflammation.
INFLUENCE OF PHYSICAL ACTIVITY LEVEL AND EXERCISE TRAINING ON TLR
Only two research groups have examined the influence of short-term exercise on TLR expression (11,12). Lancaster et al. (11) showed that 1.5 h of cycling exercise (~65% VO2max) in the heat (34°C; 30% RH) decreased the cell-surface expression of TLR1, TLR2, and TLR4 on CD14+ monocytes and also blunted the up-regulation of CD80, CD86, and MHC class II in these cell types (11). In contrast, we showed no influence of moderate-to-high intensity resistance training (nine exercises, three sets, 10 repetitions, 80% of 1-repetition maximum) on CD14+ cell-surface TLR4 expression in either untrained or resistance exercise-trained older women (12). TLR4 expression was stable for 2, 6, and 24 h postexercise but was substantially lower in the trained older women compared with the untrained older women (Fig. 3). It is possible that differences between the findings from our study and those from the study by Lancaster et al. (11) can be explained by the substantial differences in physiological stress that were applied during the short-term bout of exercise. For example, our subjects exercised at room temperature and Lancaster's subjects likely experienced considerable thermal stress. It is also possible that there is a difference in the response to endurance (11) and resistance training (12) or a difference in the response because of the differences in subjects' age.
Training and high physical activity status seem to have profound (lowering) effects on TLR4 mRNA 4 and cell-surface expression (12-14). In our initial study, we found that resistance-trained older women, irrespective of hormone status, had lower whole blood TLR4 mRNA expression than sedentary controls (4). Our subsequent studies led to the discovery that resistive-trained older women had lower CD14+ cell-surface TLR4 expression than untrained older women (12) (Fig. 3) and that physically active younger (age range, 18-35 yrs) and older (age range, 65-80 yrs) adults had lower CD14+ cell-surface TLR4 expression than sedentary, matched counterparts (13) (Fig. 4). We subsequently found that higher CD14+ cell-surface TLR4 expression in physically inactive subjects could be "trained down" to levels similar to those found in physically active subjects (Fig. 5) (14). In addition, when we grouped our subjects, irrespective of training status, into high TLR4 expressers and low TLR4 expressers, the former had significantly higher LPS-stimulated interleukin (IL)-6, TNF-α, and IL-1β production (4,12). We find these comparisons rather convincing regarding the link between TLR4 expression and inflammation, but our enthusiasm is tempered by the fact that these findings are, as yet, unreplicated by other researchers. We have also observed significant correlations between cell-surface TLR4 and LPS-stimulated cytokine production (12). In addition, we found that the daily physical activity level was modestly correlated to LPS-stimulated IL-6 (−0.279, P = 0.01), IL-1β production (−0.232, P = 0.04), and TLR4 ? (−0.234, P = 0.04) in 82 young and old adults with high and low physical activity levels (13). Thus, although we find the apparent influence of exercise and physical activity on TLR4 expression interesting, we cannot, as yet, delineate its contribution to exercise's overall anti-inflammatory effect, nor have we uncovered a mechanism for exercise training-induced lowering of TLR4. As what will become obvious in the following section, the possibilities for the latter are rapidly expanding. Clearly conceived and well-designed experiments will be required to determine potential regulators of the exercise training influence on TLR4.
MECHANISMS FOR EXERCISE TRAINING-INDUCED LOWERING OF TLR4
Regular exercise training and a physically active lifestyle have been associated with reduction in chronic inflammation (5,6,13), which may reduce the risk of developing metabolic syndrome, cardiovascular disease, and other related diseases. Despite apparent anti-inflammatory effects of exercise, the mechanism by which these effects occur is not fully understood. Based, in part, on data from our laboratory, we hypothesize that blunted TLR signaling plays a role in the anti-inflammatory effects of exercise. The TLRs are robust pattern recognition receptors that recognize a range of endogenous and exogenous substances. Endogenous ligands are HSP, ATPγs, high mobility group 1, and other yet to be identified substances (1,8), whereas the exogenous ligands are LPS, peptidoglycan, and double-stranded RNA (16). Exposure of TLR-positive cells to high doses of exogenous ligands induces endotoxic shock and cellular death; however, low-dose exposure results in tolerance (9). Tolerized cells have a blunted response to a previously introduced ligand, thus preventing disruption of normal cellular function and cell death (1,7,8). Tolerance is achieved by reducing cell-surface receptor density and/or changing pathway signaling (1,7,8). Cross-tolerance is conveyed when the receptor response to one ligand blunts the response to a second ligand. Cross-tolerance could help to explain the exercise training-induced lowering of TLR4 that we have observed, such that a damaging bout of exercise could increase both plasma and tissue concentrations of endogenous TLR4 ligands. Exposure of monocytes to endogenous ligands could induce tolerance (to endogenous ligands) and cross-tolerance to LPS. Thus, the anti-inflammatory effects of exercise may be related to the development of tolerance or cross-tolerance within the TLR pathways.
The HSPs have been implicated as a possible cross-tolerance signal for exercise-induced tissue damage because they interact with TLR2 and TLR4 (1). Pretreatment of monocytic cell lines (THP-1 cells) in vitro with HSP60 significantly decreased subsequent cellular responses to LPS and cell-surface expression of TLR4, HLA-DR, and CD86 (9)-suggesting that HSP can induce cross-tolerance against LPS. Therefore, an increased HSP may provide protection against future eccentric insults, which may be partially linked to the development of tolerance of blood monocytes and tissue macrophages. Taken together, these findings suggest that exercise and resultant tissue damage cause a release of a number of endogenous substances that may interact with TLR4. Interaction of these endogenous ligands with TLR4 has been shown to induce both tolerance and cross-tolerance (1,7,8). These effects may partially explain our previous findings with respect to cell-surface TLR4 expression and LPS-stimulated cytokine production after exercise training (4,12-14), but experiments to confirm these speculations have not been completed.
Cross-tolerance is one possible explanation for the anti-inflammatory effects of exercise; however, it is not the only plausible explanation. Changes in plasma and tissue compartment concentration of TLR accessory signaling proteins (soluble CD14 [sCD14] and LBP) or a change in their ratio (LBP/sCD14 ratio) could also help to explain the anti-inflammatory effects of exercise (15). The sCD14 is similar in structure to membrane-bound CD14; however, it allows CD14− cells, such as vascular endothelial and smooth muscle cells, to respond to ligands that signal through TLR4. The LBP is an accessory protein required for LPS binding to CD14 or sCD14 (15). It has been suggested that sCD14 and perhaps LBP play a role in the acute-phase response similar to high-sensitivity C-reactive protein (Fig. 6). More research is needed to differentiate the systemic (whole body) and peripheral (tissue compartment) effects of sCD14 and LBP (10,15).
It is possible that changes in extracellular accessory protein concentration and tolerance are not mutually exclusive but rather inclusive events associated with low-dose exposure to TLR ligands. We speculate that the anti-inflammatory effects of exercise are mediated by tolerance-induced alteration from a variety of factors (Fig. 1). More research is needed to identify other effects that may be associated with TLR signaling and tolerance after a period of exercise training.
SUMMARY AND FUTURE DIRECTIONS
Previous studies from our laboratory have shown that both a physically active lifestyle and a period of exercise training have to lower whole body inflammatory biomarkers and cell-surface expression of TLR4 (4,12-14). Nevertheless, others have not replicated our work, and we do not, as yet, have a mechanism for these exercise-induced effects. Putative mechanisms are exercise training-induced cross-tolerance or down-regulation of accessory proteins such as sCD14 or LBP.
Relationships among inflammation, TLR, and exercise training provide numerous research opportunities and several new areas of study. For example, there is an extensive array of molecules now known to influence TLR signaling and likely many, as yet, are unidentified. To our knowledge, none of these has been examined after a period of exercise training, and these studies may help explain the exercise training-induced disruption in TLR cell-surface expression that we previously observed (4,12-14).
In addition to changes in TLR cell-surface expression and signaling, there are emerging links between TLR gene polymorphisms and disease. Thus, there are a number of intriguing and important research questions that need to be answered to develop a comprehensive understanding of the link between inflammation, TLR, and exercise. It is possible that answers to these questions may provide information about exercise's consistent ability to blunt inflammation and play a positive role in inflammation-linked diseases.
The authors thank Melody Phillips, Laura Stewart, Sheri Teranishi, Kyle Timmerman, Paul Coen, and Melissa Markofski for their efforts to make the research cited in this paper possible. The authors would also like to thank Wayne Campbell, Harm HogenEsch, J. Paul Robinson, and Bruce Craig for serving as collaborators on the work cited herein. Finally, thanks to our research subjects for their efforts and good humor during these studies. This study was supported by the National Institute on Aging (1RO3AG022185-01), American Heart Association-Midwest Affiliate (Ref 0350612Z), Showalter Trust Foundation, and Center on Aging and the Life Course, Purdue University.
1. Beg, A.A. Endogenous ligands of toll-like receptors: implications for regulating inflammatory and immune responses. Trends Immunol.
2. Bruun, J.M., J.W. Helge, B. Richelsen, and B. Stallknecht. Diet and exercise reduce low-grade inflammation
and macrophage infiltration in adipose tissue but not in skeletal muscle in severely obese subjects. Am. J. Physiol. Endocrinol. Metab.
3. Chow, J.C., D.W. Young, D.T. Golenbock, W.J. Christ, and F. Gusovsky. Toll-like receptor-4 mediates lipopolysaccharide
-induced signal transduction. J. Biol. Chem.
4. Flynn, M.G., B.K. McFarlin, M.D. Phillips, L.K. Stewart, and K.L. Timmerman. Toll-like receptor 4
and CD14 mRNA expression are lower in resistive exercise-trained elderly women. J. Appl. Physiol.
5. Gielen, S, V. Adams, A. Linke, S. Erbs, S. Mobius-Winkler, A. Schubert, G. Schuler, and R. Hambrecht. Exercise training in chronic heart failure: correlation between reduced local inflammation
and improved oxidative capacity in the skeletal muscle. Eur. J. Cardiovasc. Prev. Rehabil.
6. Greiwe, J.S., B. Cheng, D.C. Rubin, K.E. Yarasheski, and C.F. Semenkovich. Resistance exercise decreases skeletal muscle tumor necrosis factor alpha in frail elderly humans. FASEB J.
7. Kariko, K, D. Weissman, and F.A. Welsh. Inhibition of toll-like receptor and cytokine signaling-a unifying theme in ischemic tolerance. J. Cereb. Blood Flow Metab.
8. Kaufmann, A, B. Musset, S.H. Limberg, V. Renigunta, R. Sus, A.H. Dalpke, K.M. Heeg, B. Robaye, and P.J. Hanley. "Host tissue damage" signal ATP promotes non-directional migration and negatively regulates toll-like receptor signaling in human monocytes. J. Biol. Chem.
9. Kilmartin, B, and D.J. Reen. HSP60 induces self-tolerance to repeated HSP60 stimulation and cross-tolerance to other pro-inflammatory stimuli. Eur. J. Immunol.
10. Kitchens, R.L., and P.A. Thompson. Modulatory effects of sCD14 and LBP on LPS-host cell interactions. J. Endotoxin. Res.
11. Lancaster, G.I., Q. Khan, P. Drysdale, F. Wallace, A.E. Jeukendrup, M.T. Drayson, and M. Gleeson. The physiological regulation of toll-like receptor expression and function in humans. J. Physiol.
563(Pt 3):945-955, 2005.
12. McFarlin, B.K., M.G. Flynn, W.W. Campbell, L.K. Stewart, and K.L. Timmerman. TLR4 is lower in resistance-trained older women and related to inflammatory cytokines. Med. Sci. Sports Exerc.
13. McFarlin, B.K., M.G. Flynn, W.W. Campbell, B.A. Craig, J.P. Robinson, L.K. Stewart, K.L. Timmerman, and P.A. Coen. Physical activity
status, but not age, influences inflammatory biomarkers and toll-like receptor 4
. J. Gerontol. A Biol. Sci. Med. Sci.
14. Stewart, L.K. Influence of exercise training and age on CD14+
cell-surface expression of toll-like receptor 2 and 4. Brain Behav. Immun.
15. Stoll, L.L., G.M Denning, W.G. Li, J.B. Rice, A.L. Harrelson, S.A. Romig, S.T. Guunlaugsson, F.J. Miller Jr., and N.L. Weintraub. Regulation of endotoxin-induced proinflammatory activation in human coronary artery cells: expression of functional membrane-bound CD14 by human coronary artery smooth muscle cells. J. Immunol.
16. Takeda, K., and S. Akira. Toll-like receptors in innate immunity. Int. Immunol.
17. Xu, H., G.T. Barnes, Q. Yang, G. Tan, D. Yang, C.J. Chou, J. Sole, A. Nichols, J.S. Ross, L.A. Tartaglia, and H. Chen. Chronic inflammation
in fat plays a crucial role in the development of obesity-related insulin resistance [see comment]. J. Clin. Invest.